[雙語翻譯]--外文翻譯--對(duì)外漢語學(xué)習(xí)者通過語義和語音進(jìn)行漢字解碼的新興發(fā)展（原文）

1、Emerging development of semantic and phonological routes to character decoding in Chinese as a foreign language learnersClay WilliamsPublished online: 25 March 2012 © Springer Science+Business Media B.V. 2012Abstrac

2、t This study examines the effects of semantic and phonetic radicals on Chinese character decoding by high-intermediate level Chinese as a foreign language (CFL) learners. The results of the study suggest that the CFL lea

3、rners tested have a well- developed semantic pathway torecognition;however, their phonologicalpathway is not yet a reliable means of character identification. Semantic radicals that correctly pertain to character meaning

4、 facilitated reaction time in semantic categorization tasks, while radicalsthathad noimmediately interpretablerelationtocharactermeaning had a strong inhibitory effect. The relative accuracy of phonetic radicals (for pre

5、dicting the whole- character’s pronunciation) did not measurably improve homonym recognition. In a lexical decision task (Experiment 3), the subjects were significantly slower in identi- fying pseudo-characters when the

6、phonological component was blurred, indicating that, despite having unreliable phonological pathways to character recognition, the subjects were still using phonological radical analysis as their default recognition stra

7、tegy; however, the author argues that the radical is likely being used for orthographic dis- ambiguation more than for phonological properties.Keywords Chinese character processing · Chinese as a second language

8、83; Dual route · Semantic activation · Phonological activationReading in ChineseProcessing Chinese: a dual route modelThe Chinese character system lends itself well to the study of reading processes. As Chinese

9、 employs a complex set of characters—each representing both a meaningC. Williams ( Shen Tan, Hoosain, Zhou Zhou, Shu, Bi, & Shi, 1999), which cannot be fully explained by the traditional non-lexical route model. T

10、hus, Weekes, Chen, and Yin (1997) proposed a Chinese- specific ‘triangle’ model containing three levels of representation—semantic, ortho- graphic, and phonological—all linked via two bi-directional pathways: a semantic

11、pathway and a non-semantic pathway. In this model, the nonsemantic pathway varies from the traditional nonlexical route in that it allows for phonological representation at both the character and the sublexical level. Fu

12、rthermore, the orthographic level of representation allows for whole character recognition (which would obviously be necessary for successful decoding of the comparably small proportion of characters which are not semant

13、ic-phonetic compounds—i.e., the true pictographs and ideographs —and also is likely to act as an early search strategy, before literacy development makes sub-processing of character radicals more efficient).Priming studi

14、es: evidence of dual routes to character decodingChinese word-recognition should be primed via presentation of semantically or phonetically related stimuli if individual characters are recognized through sublexical seman

15、tic and phonological representation. Studies have indeed shown semantic facilitation (e.g., Zhou & Marslen-Wilson, 2000). Feldman and Siok (1999) found significant semantic facilitation for characters with the same r

16、adical that were semantically related. Flores d’Arcais (1992) found interference effects between characters with different meanings, but one character’s radical was semantically related to the other character. Ding, Peng

17、, and Taft (2004) found that primes with the same radical as the target would facilitate recognition, so long as the radical was in the same position for both characters. As for phonological priming, Perfetti and Tan (19

18、98) claimed that phonological activation precedes semantic activation (see also Tan, Hoosain, & Siok, 1996). Zhou and Marslen-Wilson (2000) found phonological priming effects as well, but only for prime durations of

19、200 ms. Shen and Forster (1999), argued that phonological priming in Chinese is task-dependent—exempli gratia, that naming studies would predispose subjects to phonological interpretations, and that semantically dominant

20、 tasks would encourage a semantic processing strategy. Williams and Bever (2010) followed this line of reasoning, and proceeded to show that subjects would demonstrate strong semantic facilitation effects in a semantic c

21、ategorization task and strong phonological facilitation effects in a homonym recognition study. Wu and Liu (1997) found that phonological activation only follows the encoding of both the phonetic and the semantic radical

22、s. Seidenberg (1985) only found facilitation from the phonetic component in low frequency words (which makes sense as lower frequency correlates with higher phonological transparency on the part of the phonetic component

23、. Taken as a whole, the evidence seems to indicate that both the semantic and the phonetic components of semantic-phonetic character compounds can serve as focal points for lexical searches.Emerging development of charac